Population Dynamics in ARIEL Robotics Systems Featuring Embodied Evolution via Spatial Mating Mechanisms
For researchers designing spatially-embedded evolutionary algorithms, this work provides empirical constraints showing that common spatial mechanisms (e.g., proximity-based mating, density-dependent death) can destabilize populations or reduce fitness.
The paper investigates how spatial structure affects evolutionary dynamics in a simulated robot population using HyperNEAT controllers, finding that spatial selection mechanisms introduce trade-offs such as unstable dynamics, fitness decline, or bistability, with only deterministic fitness-based selection maintaining stability.
We present a Spatially Embedded Evolutionary Algorithm where robot individuals exist in a physically simulated 2D environment, must navigate to encounter potential mates, and compete for survival under various spatially-aware selection pressures. Using HyperNEAT evolved neural controllers for ARIEL gecko-inspired quadrupeds in MuJoCo, we investigate how spatial structure fundamentally alters evolutionary dynamics. Our experiments show a modest 4.9% difference in peak fitness between proximity-based and random pairing possibly within stochastic variation while combining spatial parent selection with stochastic death selection produces unstable population dynamics. We discover a continuous phase transition in energy-based selection experiments, with critical zone count separating extinction-dominated and explosion-dominated regimes. Our density-dependent death selection mechanism achieves 97% completion rates but causes fitness decline, revealing a fundamental dilemma where decoupled mechanisms produce bistable dynamics, positively coupled mechanisms create counter-selection pressures, and only deterministic fitness-based selection maintains stability. These findings provide important constraints for future spatial EA design.